End-to-End Evaluation for Low-Latency Simultaneous Speech Translation

TL;DR

This paper introduces a comprehensive framework for evaluating low-latency speech translation systems in realistic scenarios, enabling fair comparison of different approaches including end-to-end and cascaded models.

Contribution

It presents the first end-to-end evaluation framework for low-latency speech translation, considering segmentation, runtime, and quality in real-world conditions.

Findings

End-to-end framework enables realistic evaluation of latency and quality.

Comparison of models with output revision and fixed output.

State-of-the-art systems evaluated under consistent conditions.

Abstract

The challenge of low-latency speech translation has recently draw significant interest in the research community as shown by several publications and shared tasks. Therefore, it is essential to evaluate these different approaches in realistic scenarios. However, currently only specific aspects of the systems are evaluated and often it is not possible to compare different approaches. In this work, we propose the first framework to perform and evaluate the various aspects of low-latency speech translation under realistic conditions. The evaluation is carried out in an end-to-end fashion. This includes the segmentation of the audio as well as the run-time of the different components. Secondly, we compare different approaches to low-latency speech translation using this framework. We evaluate models with the option to revise the output as well as methods with fixed output. Furthermore, we directly compare state-of-the-art cascaded as well as end-to-end systems. Finally, the framework allows to automatically evaluate the translation quality as well as latency and also provides a web interface to show the low-latency model outputs to the user.